doing the public good: ethics, ethnography and policy research
TRANSCRIPT
4 Aug 09
1
Doing the Public Good: Ethics, Ethnography and Policy Research
Abstract for 4th International Ethnography Symposium
Pam Carter, Keele University [email protected]
1. Introduction This paper adds to contemporary debates about research governance using data
from my PhD ethnographic research project that was subject to the UK Research
Governance Framework (RGF) for Health and Social Care (DH, 2005). I discuss
the ethics of “getting in and getting on” including the bureaucratic process of
gaining ethical approval via the local Research Ethics Committee (LREC) and the
ethical dilemmas I encountered during my research practices of observation,
transcription, analysis and writing up. I outline some of my ethical dilemmas,
produce a “confession” of an ethical misdemeanour and raise the question of
how effective governing codes are in practice. I conducted much of my PhD
research in a state of uncertainty and confusion and I am using this paper as an
opportunity to reflect and attempt to surface some of the more or less
unconscious decisions I took to resolve ethical dilemmas. Following a discussion
of contemporary research governance debates, I provide some background on
my research, detail the arduous but crucial process of obtaining ethical approval,
before going on to reveal what happened next. I show how at various moments
during my research endeavour I experienced research governance as
constraining, as enabling and as ineffectual.
2. Research governance concerns for practising ethnographers
Shaw et al (2009:918) call for further research on research governance with a
focus on “…examining how the tensions and contradictions that are
foregrounded in health research and governance are ‘best’ played out.” This
4 Aug 09
2
paper follows up their recommendation, not by presuming that my experience is
any example of “best practice” but with an aim of illustrating how research
governance in practice affected my ethnographic research on childcare policy
implementation. Critical policy research has a long tradition of an ambition to
“speak truth to power” (Haas, 2004) which appears to positions policy
researchers as ethically motivated, on the side of truth against policy makers who
pursue their own, often corporately and/or politically aligned interests. Ideas
about research being in the public interest and for “society in general” (SPA,
2009) are prevalent in ethical codes (ESRC, 2005, DH, 2005, BSA, 2004).
Sometimes, however, notions of research carried out in “the public interest” can
be theoretically and politically naïve (Gillies and Alldred, 2002). Ideas about the
public good have come under increasing attack from neo-liberal theories, policies
and practices (Burawoy, 2005, duGay, 2005). This has been accompanied by an
uneven trend of a consumerist discourse alongside a decline in deference
towards traditional forms of authority, including the professions (Newman and
Clarke, 2009). These authors refuse a simple understanding of a “general
public”, analysing how contemporary governance practices produce, constitute or
call up a variety of publics from “citizen-consumers” to local communities to
stakeholders with special interests. Foucault’s (1980) analysis demonstrates the
intimate relationship between power and knowledge problematising ideas of
interest-free, neutral and objective scientific research. Feminist research has
been influential in deconstructing the “bounded, rational and autonomous”
Western model of the individual subject (Alldred and Gillies, 2002:146).
Ethnographers are by now (or ought to be) familiar with debates about the
cultural / postmodern turn (Clifford and Marcus, 1986) which can result in what
Bailey (2001) terms “veriphobia”.
These epistemological and ethical debates are surfacing in debates about the
legitimacy of Research Governance (Aldred, 2008, Shaw and Greenhalgh, 2008,
Calvey, 2008). Ethical codes and frameworks pay attention to informed consent
4 Aug 09
3
as a key ethical principle that is designed to minimise harm but as Hoeyer et al
(2005) discuss, the disciplines of anthropology and medical research have quite
different traditions that can mean divergent understandings of harm and
beneficence. The Research Governance Framework (2005) that applies to all UK
research carried out within the NHS (heareafter RGF) acknowledges harm as a
relative concept to be weighed against beneficial outcomes of research. Local
Research Ethics Committees (LRECs) are charged with weighing up potential
harm to research participants against the public benefit of research. Aldred’s
ethnography “studied up” elites (Aldred, 2008). She discusses the limitations of
considering ethics purely in terms of the “researcher-researched dyad” and notes
that harm may be done to organisations, not just to individuals.
Researching the powerful situates the ethnographer critically as the stance of
researching “up” already implies recognition of power. The British Sociological
Association recognises such issues in the section of their statement of ethical
practice but this statement explicitly states that the statement is not to be used
as a
“set of recipes for resolving ethical choices or dilemmas … but recognises
that it will be necessary to make such choices on the basis of principles
and values, and the (often conflicting) interests of those involved.” . (BSA,
2002:3) . 1
Dingwall, (2006) following Haggerty (2004) takes issue with what he sees as
inappropriate “ethics creep” as research governance imposes control over
increasing areas of academic inquiry. Richardson and McMullan’s (2007) survey
research of academics’ opinions of the RGF found that some respondents had
avoided researching within the NHS altogether and the authors cite Bently and
Enderby’s (2007) finding that applications had dropped by 40% .Other
respondents had resorted to “gaming” to avoid the scrutiny of NHS ethics
4 Aug 09
4
committees by classifying studies as “audit” rather than research. Edwards and
Mauthner point to the risk of assuming that “an ethics committee sanctioned
project is by definition an ethical one.”(Edwards and Mauthner, 2002:p.18).
Whilst understandable as a response to perceived over-regulation, posturing and
pretence by researchers is surely not in accordance with the spirit of ethical
codes. Rather, the practice constitutes what Dingwall (2006:51) terms
“ceremonial conformity”.
Hedgecoe (2008) carried out an ethnographic study of several LRECs. His study
reports on the process of considering research applications for ethical approval
rather than following up subsequent research and governance practices to see
how the approval process impacts on the conduct of ethnography. In contrast I
focus on my practical ethical quandaries and how these were mediated by
research governance as an example of how “those operating outside or on the
boundaries of dominant epistemological frameworks struggled with tensions
between systems.” (Shaw et al, 2009:917). Although the RGF (DH, 2005)
appears to recognise a variety of methodologies, section A10 of the application
form asks for:
“a full summary of the purpose, design and methodology of the planned
research, including a brief explanation of the theoretical framework that
informs it. It should be clear exactly what will happen to the research
participant, how many times and in what order. Describe any involvement
of research participants, patient groups or communities in the design of
the research.”
This question implies a hypothetico-deductive model of research privileging the
clinical, randomised controlled trials frequently used in medical, rather than
sociological, particularly ethnographic research. This presents ethnographers
with the task of fitting the square peg of open ended, trust-based ethnography
4 Aug 09
5
with its emergent theorizing into the round hole of the RGF with its expectations
of a predictable research design, a clearly specified theoretical framework and a
contractual insistence on fully informed consent from research participants
(Murphy and Dingwall, 2007, Richardson and McMullan, 2007). I suggest that,
whilst ethnography as a method is not necessarily more true, more just or more
radical than other data gathering techniques it does require more personal
investment from the researcher, relying as it does on “close to the bone”
encounters and so it can be more risky and unpredictable (Back, 2002, Willis and
Trondman, 2000).Having outlined some of the debates about research
governance, the next section presents empirical data on my own research
governance encounter.
4. Ethical Dilemmas encountered in one piece of ethnographic policy research
In this section I show I negotiated ethical approval, describe the process of
gaining access to a research site then how I subsequently dealt with data
generation and analysis. Throughout these phases of my research, I show how I
managed ethical dilemmas and interacted with the requirements of research
governance
Context
My PhD study analysed the local implementation of the UK Government’s
childcare policy in England. I started my research with ambivalence about
whether the increasing commodification of childcare is a good thing, and a
concern for whether the emancipatory potential of the childcare strategy for
liberal feminists might lead to unintended consequences of workfare for working
class mothers living in poverty (Lewis, 2006). Besides sampling some relevant
policy conferences and a council cabinet meeting, I observed a Community
Partnership network operating in the quasi-public arena of “partnership”
4 Aug 09
6
governance, with governance arrangements initially unclear about what
constituted commercial interest, what ought to be in the public domain and what
might remain a personal or private matter. (Newman 2009). I carried out an
ethnographic study of the complex governance of childcare as the UK public
policy of Sure Start Children’s Centres was being implemented by local
authorities and their partner organisations. These included NHS Trusts as part of
Children’s Trusts and private sector nursery businesses working in “partnership”.
The research was sociologically, not medically oriented. Rather than interviewing
or observing policy beneficiaries, I “studied up”. For the most part I observed
people (usually in middle management positions) who made decisions that
affected the lives of children and families. My research enquiry was informed by
my experience of working as a local government practitioner, committed to a
public service ethos, by my feminist sensibility 2and interest in how childcare
policy related to the UK child poverty strategy. When researchers define a
“context” for their research, they have framed reality and thus constructed,
however vaguely at the outset, an ontological perspective and so ethical issues
can begin to be anticipated and considered. (Edwards and Mauthner, 2002)
Getting In: Informed Consent
I was required to complete a University “Research Protocol Outline Form” in
addition to the main NHS Research Ethics Committee application form with some
compatibility between the forms enabling cut and paste. Both forms contain a
section asking for a “scientific justification for the research” requiring answers to
the questions “What is the background? Why is this area of importance?” which
“must be in language comprehensive to a lay person.” Constructivists might
regard one person’s social significance as another’s dross piece of “so what”
research and the language of social theory can be notoriously abstruse. I gave
what seemed at the time and on reflection at a distance still seems a
2 There is no single version of feminist theory – my own position holds in tension a recognition of the value of equal opportunities liberal feminism, that might be useful in the here and now, with a more future oriented radical vision of a redistribution of the sexual division of labour and ethic of care.
4 Aug 09
7
comprehensive honest answer, detailing my view of the policy context. I do not
believe that ethnographers should adopt a defensive position with respect to their
methodological expertise. Hannigan and Allen (2003:692) point out that
“Researchers, for their part, need to be sensitive to the interpretative work
that inevitably goes on in REC decision-making, and ensure that the
information contained in their applications is unequivocal. “
This negates the idea that language is always subject to interpretation. One
justification for ethnography as an appropriate methodology for policy research is
a technical argument for the inadequacy of experimental or survey methods in
“getting beyond the official line” (Duke, 2002) and for understanding how policy
implementation happens within communities of practice (Lave & Wenger, 1991).
There is a body of research identifying a policy “implementation gap” (Barrett,
2004) that supports the argument for ethnographic research. I suggest that
qualitative researchers (and perhaps ethnographers in particular) possess
rhetorical skills and on reflection I see how I deployed these in constructing a
justification for my research in my ethics application (Golden-Biddle and Locke,
1995).The section of the application form “Scientific Background” asks “Have all
existing sources of evidence, especially systematic reviews been fully
considered?” I stated that I was aware of systematic reviews in the fields of
“health inequalities, social inclusion, the effect of day care on children and
welfare to work policies in the US.” but as I look back I realise that I did not
provide citations. This was not queried. I made my case that my research
“Through close-up observation, interviews and documentary analysis, new
information will be generated about the meaning that a variety of policy
actors attach to parenting, work and family support and the processes
through which such meanings may shift or stabilise as the policy is
implemented.”
4 Aug 09
8
The challenge of communicating in lay language at this early stage in my
research was considerable but this did help me to formulate my ideas. Shaw et al
(2009:917) draw attention to “…how decision-making tools highlight some
aspects of experience but silence others” and they note that the iterative nature
of qualitative research is silenced. My application did not silence these unique
aspects of ethnography. In the section “Detailed plan of investigation”, despite
being directed “It should be clear exactly what will happen to the research
participant, how many times and in what order” I stated “The design is intended
to be flexible in order to respond to opportunities for further data collection and
any practical constraints that arise during the course of the fieldwork.” In the
section “Detailed Plan of Investigation” I followed the suggestion given on the
form to use the sub-heading “Theoretical Framework”. I set out my “theoretical
interests” which I explained were
“… informed by cultural studies and feminist theory, … Fischer’s (1993)
work on policy argumentation, and on a discursive tradition within
qualitative research that focuses on the ways in which language mediates
reality (Fairclough, 1992, Levitas, 1998).
My protocol was peer reviewed by two academics but this was not a blind
process. The form asked me to provide the names of external reviewers and
state if I had any relationship with them. I was able to state “none” but the
reviewers were colleagues of the academic acting as my supervisor at that time
and may have been inclined because of their friendship to give my research a
favourable opinion. I completed the online COREC 3 form and attached a host of
documentation such as my own and my supervisors’ CVs, a 4 page research
protocol, an information sheet about my research, evidence of funding, a
template detailing sponsor responsibilities, an invitation letter to participants, and
a form I designed to obtain informed consent. The University Independent Peer
Review Committee scored my application as “grade 2” and requested:
3 Central Office for Research Ethics Committees see http://www.nres.npsa.nhs.uk
4 Aug 09
9
1. More detailed explanation of the background including a clearer rationale
for undertaking this study
2. More information regarding the structure of the interviews
3. Consideration to the use of the standard COREC consent form.
I responded positively to the first point but pointed out that “It is not possible to be
more specific at this stage about the study population. Early enquiries suggest
that local authorities are struggling to engage health partners in this policy
[Children’s Centres] …” I hastily produced a very loosely structured topic guide to
address the second point stating that:
“I have discussed this with my supervisor who considers this to be
adequate at this stage but we would like to stress that it is likely to be
amended and refined as the fieldwork progresses.”
On the third I pointed out that:”
I have reviewed the standard COREC consent form but I believe that the
references to G.P. and medical records are inappropriate for this study
which does not aim to collect any medical data. “
Rather than complying with the medically dominated framework, I argued my
case and I offer this as a recommendation to other social researchers although I
acknowledge that LRECs are not necessarily consistent in their procedures
(Hannigan and Allen, 2002). My response resulted in a grade 1 “proceed without
any revision”. To know that my protocol had been peer reviewed was a boost to
my morale at this stage of my PhD but this only got me as far as the University
ethical approval. I still had to gain approval from the Local Research Ethics
Committee.
4 Aug 09
10
I attended the LREC with my supervisor to answer questions. A lay member of
the committee asked what I felt was the most intelligent, appropriate question
about the political nature of the research. I agreed that I would be encountering
“political” situations. The question of politics did not seem to have occurred to the
other committee members, who included medical researchers. We had a
dialogue about my plan to observe meetings. To ensure informed consent from
people who would be attending the meetings that I wanted to observe, I was
advised to hand out my information sheet, withdraw from the meeting to allow
time for a discussion and to repeat this on any subsequent meeting where the
group members may have changed. I found this impossible to comply with in
practice.
Getting In: access to data
The research governance process took 6 months from drafting the application to
finally obtaining ethical approval. During this time I approached a local authority
to request permission to observe policy implementation within their organisation.
The “ways in” to the organisations are porous in some instances but can appear
closed and tightly controlled in others (Bryman, 1988). I contacted a Council that
was within travelling distance and I guaranteed anonymity to the local authority
that agreed to act as sponsor of my research. In exchange for guaranteed
anonymity I negotiated permission from a local authority to research their
implementation process and immediately encountered ethical dilemmas around
gatekeeping. 4 I met with a senior officer at a County Council – I’ll call him “Ted”
and negotiated permission to carry out my PhD study. We discussed the “quid
pro quo”. Ted expressed his interest in the theme of governance. The meeting
was friendly and fruitful – permission was granted in return for guaranteed
anonymity and a summary report of the research to be presented at some time in
the future to (some unspecified part of ) the local authority. To protect the
Council’s anonymity I requested of an NHS R&D Manager that details of my
4 See Miller and Bell (2002)) for a detailed discussion of gate-keeping
4 Aug 09
11
research application be excluded from the National Research Register as this
would identify the local authority.
Getting On
My anxiety about the PhD process and naïve novice researcher status led me to
begin pursuing all practical possibilities for data collection in a somewhat
desperate fashion. Sometimes I drew on my existing social capital to request
interviews (Edwards et al 2004). Snowballing sampling can have ethical
implications as frequently I was asked whether I had already spoken to named
individuals. I generally fudged a reply or lied in order to protect confidentiality.
One manager of a Children’s Centre wanted reassurance that I had some
existing research experience before he agreed to be interviewed. I explained that
I understood his concerns about allowing inexperienced research students
access for their research as I had previously faced this same dilemma myself.
This demonstrates the ethical dimension of competence which poses an
inevitable problem for students. The ESRC Research Ethics Framework
(2005:16) recognises the dilemma but in my case my student status necessitated
the same type of ethical review procedure as any other more qualified or
experienced researcher would be subject to.
Generally I found the process of interviewing individuals less emotionally
stressful than being in meeting situations where I made observational field notes.
Interviews are relatively structured, predictable interactions with social
conventions and professional props such as a tape recorder that structure mutual
expectations. I felt safer, on surer ground compared to the unpredictability of
open-ended observational fieldwork where I used a pen and notebook to write
fieldnotes. The advantage of this recording method was that most other people at
the meetings also made notes so my activity was unremarkable. Sometimes I
was conscious of people looking over my shoulder to see what I was writing or
seeming surprised at the volume of my notes. The amorphous nature of the
4 Aug 09
12
partnership with its fluid membership and “round tables” meant that I did not often
stick out like a stranger. Sanger (1996:35-36) has useful advice on “being a
chameleon” including “Emulate the norm in the activity being observed” and “Try
not to make eye contact”. I found it relatively easy to gain access to one local
network, partly because membership was defined loosely. One way in which I
interacted with people responsible for policy implementation was to adopt a role
of an intellectually curious outsider seeking knowledge from experts. There may
have been some dissembling here as I was familiar from my own past work
history with much of the activity I was researching. This resonates with the “…
risk of feeling, and indeed actually becoming, ‘phoney’ and ‘inauthentic’ when
“doing rapport” that Duncombe and Jessop (2002) discuss as an ethical
dilemma. “Getting on” was a matter of demonstrating a keen interest in and
respect for the demands of managers’ policy implementation responsibilities. I
naively assumed that I had been given permission to “access all areas”.
However, when I contacted managers more senior than Ted in the authority to
request interviews with them, my e-mails were intercepted by their secretaries
whose response was that “Ted is dealing with this policy implementation and can
tell you everything you need to know.” The fact that the policy is being delivered
within complex governance arrangements influences access as I interviewed
several people working outside the Council hierarchy in the voluntary sector and
so outside RGF jurisdiction.
Attendance at meetings seemed to comprise a core membership as well as
several people who seemed to drift in and out with no opportunity for me to
establish contact or rapport. Whenever the opportunity arose (generally during
the round robin introduction part of a meeting) I introduced myself as a
researcher, flagged up my information sheet with an offer to give one to anyone
who wanted a copy, I guaranteed anonymity and offered to answer any questions
regarding my research. Often I found myself making use of the credentials of the
LREC to give reassurance that my research was being conducted ethically.
When introductions did not take place, it was because the group had begun to
4 Aug 09
13
expect that, after several meetings, people would know one another or that
newcomers were expected to catch up as there was business to get on with. It
felt unethical to disrupt proceedings and I frequently found myself making moral
judgements “on the hoof” about my overt / covert researcher status, weighing up
potential harm against benefits and assessing risk to myself and to others. I am
not suggesting that I was in physically dangerous situations; the nature of my
research was not that I was asking questions about topics more usually
considered private or sensitive such as sexuality. The sensitivities arose from
the political nature of observing a group, talking to individuals about their role in
the group and about asking for views and opinions on “official” policy while
negotiating relationships of trust. My partial insider knowledge did not necessarily
automatically gain me entry or automatically admit me once I had my ethical
approval, or my “anthropologist’s visa”. 5 Opportunities to observe “policy in
action” were contingent, dependent upon finding out about policy implementation
meetings in the first place, negotiating access, building credibility and trust or
sufficient rapport to be invited back (Buchanan et al 1988). Sometimes I
requested permission in advance from the Chair. On at least one occasion I
risked turning up uninvited and my researcher status went practically undetected
until I found myself about to be co-opted onto a management group. I was called
upon to explain why I wasn’t voting and explained my independent status. Where
sub-group meetings were small, the intimate nature of the group affected my role
– I was more open to scrutiny and “stuck out” and so chose not to attend some
meetings where I might risk spoiling the rapport I had developed in the larger
partnership. I requested private interviews with most of the people I observed. To
maintain confidentiality, I had to ensure in subsequent meetings that I gave no
indication that these private interviews had taken place. I am confident that I
have maintained confidentiality, (despite meeting a professor at a Christmas
party who seemed determined to try to get me to break my code of ethics,
pressurising me to reveal the name of the Council in my study). Some people
declined to be interviewed but I continued to observe them. I believe that the
5 This phrase was used by Teresa Smith at a Social Policy Association Postgraduate event I attended.
4 Aug 09
14
course of meetings and so the data I gathered was not altered by my presence.
This messy reality is considerably at variance with the idea of a planned study
design but not with an iterative research design as envisaged and presented by
me in my protocol for the ethics committee.
Having gained some distance with the passage of time I can make sense of
some of the feelings of discomfort I felt in the field. I felt like a gamekeeper turned
poacher. I had checked with some past work colleagues whether they felt that it
would be at all realistic for me to conduct observation and they replied that yes,
so many people working in the public sector are used to being surveilled,
monitored and judged by the likes of SCIE 6or Ofsted7 that this was practically a
commonplace experience (Law and Mooney, 2007). It was only later that I
understood the risk of entering the field and being perceived as an inspector
(Clarke and Hall, 2008). This “placing” or framing of my researcher role as
outsider sometimes militated against my attempt to behave as a participant
observer. I learned that detaching myself in a physical sense from the main
meeting could be interpreted by other participants as either self-exclusion, and
they felt sorry for me, or I was positioned in the “judge” role which made them
uncomfortable so on most occasions I participated alongside them in meetings.
On one occasion I found myself the only member of a group willing to draw a
picture of a pram so I duly made my mark on a flipchart. Early on in my
research, following my attendance at several Council consultation events I
presented Ted with a report analysing what I had found in terms of “cultural
attitudes to change.” I had one further meeting with him and then contact went
“cold” which I experienced as a relief. Most of my observations of the Community
Partnership were overt but other associated policy meetings I researched
covertly.
6 Social Care Institute for Excellence 7 Office for Standards in Education
4 Aug 09
15
Covert observation
I opportunistically sampled several free national and regional conferences and
“summit” meetings that took place outside of the local authority. Spicker
(2007:112) maintains that certain ethical constraints are applied inappropriately
to the field of public policy. He argues for a distinction between public and private
spheres such that
“In a democracy, if someone is functioning in a public role, that person is
subject to public examination and criticism in that role whether they like it
or not.”
In my thesis I quote senior civil servants and Government Ministers. I attended a
meeting to discuss child poverty that was held at the Treasury and a Respect
Agenda conference held at a hotel in London. At one of the meetings a Minister,
in an apparent attempt to put participants at their ease, announced from the
podium that “there are no press here today”. The implication that the meeting
was “off the record” made me wince but I remained silent, continuing to take
notes. In contrast to the medically dominated NHS Research Governance
Framework’s insistence on informed consent, the ESRC ethical framework notes
that for studies of, say, crowd control, it is impractical to secure informed
consent. It also recognises a plurality of research methodologies, that research
designs may be iterative and that in some contexts “covert observation is
necessary and warranted” Although “it is only justified if important issues are
being addressed and if matters of social significance which cannot be undertaken
in other ways are likely to be discovered.” (p.21). I did not conceal my research
student status when booking a place and always carried my information sheet.
Practically, in a large audience, my presence as a researcher was fairly covert.
However, I believed that these meetings discussing public policy constituted a
“public realm” and so I could legitimately record my observations. In my thesis
(Carter, 2009) I write “The Minister, summing up, stated that “although people
4 Aug 09
16
living in poverty have not been invited here today, nevertheless they have been
spoken about very respectfully.” In my analysis, I situate the Respect Agenda in
relation to the child poverty strategy, childcare and welfare reform. I analyse the
disjuncture between a discourse that emphasises public participation in policy
making and the semi-closed nature of a meeting where the invitation specified as
a dress code (“normal business attire”) and that proceeded with the assumption
that only professionals rather than members of the general public were invited to
the meeting. I believe my recorded observations of the meeting, theorised and
analysed, become significant in the context of my research and therefore justified
covert observation. To me it seemed ethical to expose this aspect of public policy
implementation.
5. The ethics of transcription, representation and trust
I promised all interviewees a transcription of their interview believing that this
might mediate otherwise unequal power relationships between me and
interviewees. The Local Research Ethics Committee LREC had insisted that I
offer a two staged informed consent process so that people signed to document
whether they agreed to be interviewed, whether the interview could be taped and
then signed a second form to say whether they were happy for me to use
verbatim quotations. I was not given a rationale and was too anxious to proceed,
unclear at the time how I might argue against this and, conscious of the
committee’s power of veto, I did not question this decision. Initially I obtained
signatures from interviewees once before I switched the tape on, then I
introduced the second form which dealt with use of verbatim quotations when I
had completed the interview and switched off the tape. Technically I was
complying with the LREC requirements placed on my research. However I
realised after several interviews that people would be able to give a more
informed consent to the use of their verbatim words after they had read their
transcript. I found that several people read their transcript and chose in writing
4 Aug 09
17
not to give me permission to quote them verbatim. I felt unable to query this
because of the risk of applying undue pressure but I suspect that some people
would have found it an easy choice to “play safe” and tick the “no” box. I
understood this to mean not that I could not include their interview transcripts in
my data analysis as they had already signed a Part One consent form but it did
mean that I missed the opportunity to represent some “juicy quotes”. (Schostak,
2006:122). Buchanan et al(1988) regard this form of censorship as innocuous.
They gives the example of cleaning up data to suit an organisation removing a
colloquial phrase, “he has got us by the short and curlies…” substituting
managers’ alternative phrasing “… the company is dependent on his goodwill…”.
He suggests that meaning was not altered but I would argue that colloquial
metaphors can be very revealing and I relied heavily on metaphor and other
tropes in my analysis. I transcribed non-verbal communication so that I could
analyse hesitancies, gaps and silences for what they communicated. (Alldred
and Gillies, 2002). In a couple of instances I felt the process of consent to use
verbatim quotes worked well and dialogically. Dingwall (2006:56) distinguishes
between the “fetishization of consent” governed bureaucratically through
contractual arrangements and:
“… the construction of a customized relationship between researcher and
researched, where the researched are offered explanations tailored to
their level of understanding and concerns, not presented with legalistic
formulae that require an advanced education to be intelligible…. We
cannot force our informants to provide information. We depend on their
co-operation and goodwill – but these ends are not served by the ESRC’s
demand for written evidence of a contract of consent. Contracts are
designed to manage adversarial relationships. We cannot function in
conflict with our informants.”
Rather than view the process as an adversarial contract, I attempted to manage
the ethical dilemma of informed consent respectfully. One interviewee asked me
4 Aug 09
18
to change their gender and their job title to protect their anonymity and I did so as
this did not affect my analysis. Two interviewees diverged from the options
indicated on the consent form and told me that they would prefer to see their
quotations in context before giving me their permission. This meant I was writing
up my thesis and including quotes from their transcripts without knowing whether
permission would be granted. I proceeded with the pragmatic but possibly
unethical assumption that if I did not gain their consent I would simply
paraphrase them rather than use their words verbatim. Fortunately I did not
resort to this deceit. Having drafted my thesis, I faced the choice of how much
context to give them to inform their decision to give consent or not I tried to put
myself in these peoples’ shoes to try to understand from their perspective why
they might be unwilling for me to use their words. One interviewee had been
quite critical of elected politicians whom she worked closely in partnership with
and so I guessed that she may have felt awkward or embarrassed and perhaps
wary of how secure my promise of confidentiality and anonymity could be. I sent
her a section of my thesis that incorporated her critical comments and included
similar quotes from other people as well as reference to theory to support my
argument about complex forms of governance that were being enacted. I
reasoned that perhaps she would feel reassured in not being a lone voice and
hoped that the theory and my analysis might make some sense to her. The other
interviewee’s transcript contained comments I wanted to use about her
employing organisation being a feminist organisation which was not reflected in
the organisation’s “mission statement”. Again I had to decide how much
surrounding text to extract from the thesis to inform her decision about whether
or not to grant me permission to quote her. I can only surmise that she was
satisfied that I had made reasonable use of her words and not abused the trust
she placed in me in granting me the interview and the authority to analyse her
words and her situation. This dialogic process happened outside the LREC
governance and yet felt more meaningful and ethical as it was dependent upon
trust and dialogue rather than a standardized research governance format.
4 Aug 09
19
In another instance I confess I caused harm and this confession entails an
embodied, situational approach to research and to ethics (Mauthner et al,2002) I
interviewed two local authority Members together at their home. I felt the
interview went well. In fact they were very kind to me. I turned up at their house
having just had a not too serious car accident. I was shaken and they sat me
down with a cup of tea and made me feel welcome. I felt the interview proceeded
well and was pleased with the recorded data. I posted the transcript to them and
received a response to the effect that they wished to withdraw their permission
for use of the data. They sent me a handwritten note to the effect that they felt
abused and would never take part in research of this nature again. The Social
Policy Association guidelines on research ethics states that:
“Researchers have a duty to maintain the good standing of the discipline
through the ways in which they conduct research and thereby the way in
which social policy research is regarded.”
I offer my cautionary tale as a means of educating future researchers about the
ethics of transcription and in the hope that I have not brought the discipline of
social policy research into disrepute. I have destroyed the transcript since the
complaint but I shall explain how I constructed it. When I was gathering data I
was uncertain how I would subsequently be analysing it and thought I might give
myself the option of carrying out conversation analysis if I produced transcripts in
an appropriately detailed format. I transcribed the interview myself. The
Councillors spoke in a local dialect that I shared and sought to reproduce. I did
not correct grammatical errors believing that it would be more honest to send a
transcription in the same format that I might subsequently want to reproduce
verbatim quotations from – i.e. colloquial speech. I received a written note from
these interviewees to the effect that they felt I had abused their trust. I replied
apologising and offering a meeting with my research supervisors but heard no
more. Whilst their reaction could be rationalised as a lay misunderstanding of the
transcription process and of the purpose of qualitative research, I personally felt
4 Aug 09
20
ashamed of my conduct. At some level I had been aware that I wanted readers of
my thesis to infer from quotations aspects of the interviewees’ identity such as
their working class location. Theories around representing marginalised
populations advocate allowing research participants’ voices to be heard.
(Schostak, 2006). Voice is essentially different than the written word in that non-
verbal communication, tone, dialect and inflection convey more meaning than
“mere” words Alldred and Gillies (2002). I became immediately aware of the
harm I had caused in representing these interviewees as undereducated. Several
other interviewees remarked on how often they said “y’know”, er or um but I felt
that their cultural capital gave them the psychological resource to cope with a
transcript that only made them appear different than they might otherwise have
represented themselves in their own form of written words. I also pointed out that
I included my own words in their transcripts so that when I asked questions I
included my own hesitancies and grammatical errors of speech. (Hoeyer et al,
2005:1746) write: “The psychological reaction to seeing oneself and one’s ideas
described, objectified and relativised, is difficult to predict and, thus, difficult to
prepare for”. Corden and Sainsbury (2005) conducted a study to explore the
issues of verbatim quotes. This resulted in equivocal findings so that ethical
dilemmas remain around transcription and representation. I still feel ashamed
that I caused harm. .
I volunteered to present interim findings back to the CLP whose meetings I
attended over the course of several months but the offer was not taken up. I feel I
have partially fulfilled my obligation to report back to research participants. On
the other hand, I cannot possibly claim that my research has made any
difference to the people I met in the field.
4 Aug 09
21
6. Being Monitored (or not) Governing practice included sending an annual report. As discussed in my
“confession”, I felt that I had experienced a “serious adverse event” but this was
not as constructed by the RGF in which a SUSAR (Suspected Serious Adverse
Reaction) carries medical implications.8 “Harm” seems to be generally
conceived of within the DH RGF lexicon as bodily harm or emotional distress
caused by the immediate consequences of taking part in research which may
involve medical interventions. For example, under Health and Safety, section the
RGF states “Research may involve the use of potentially dangerous or harmful
equipment, substances or organisms. “2.2.8 Some research may involve an
element of risk to those participating in it. If there are any risks to participants, the
risks must be in proportion to the potential benefit. Under the section headed
“Ethics” 2.2.8 we find “Risks, pain or discomfort must always be kept to a
minimum, and explained clearly both to the relevant research ethics committee
and to participants. It must always be explained whether there are arrangements
for compensation in the unlikely event of nonnegligent harm. Alldred (2008) says
a key question is who regulates but a further research question I propose is how
effective is the regulating? The harm that the interviewees were complaining that
I had caused them remained outside the LREC’s purview, partly because of their
medical bias.
.
In an annual progress report to the LREC in the “amendments” section which
asks” Have any substantial amendments been made to the trial during the year?”
I stated in italicized block capitals “this is not a trial” and wrote disingenuously:
8 “An 'adverse reaction' is any untoward and unintended response in a subject to an investigational medicinal product which is related to any dose administered to that subject. An adverse reaction is “serious” if it: (a) results in death (b) is life-threatening; (c) requires hospitalisation or prolongation of existing hospitalisation; (d) results in persistent or significant disability or incapacity; (e) consists of a congenital anomaly or birth defect. A 'suspected serious adverse reaction', therefore, is any event which is suspected of meeting the above criteria
4 Aug 09
22
“my research has had to respond to the exigencies of fieldwork” without
describing what this process was. I was covering my back. An astute reader of
my report responsible for monitoring my research picked this up and asked for
clarification at which point I responded honestly as follows:
I recently completed a monitoring form and stated under ‘deviations’ that
‘The protocol has needed to adapt to the contingencies of researching
policy implementation in action.’ I realise that this statement is ambiguous
so I wish to clarify. I have met with my main supervisor recently
specifically to reflect on and discuss the ethical implications of my
research. I have returned to the COREC guidance that references the
ESRC (my funder) which also has its own very useful ethical guidelines.
My original protocol stated in section A10 that “The design is intended to
be flexible in order to respond to opportunities for further data collection
and any practical constraints that arise during the course of the fieldwork.”
The LREC approved this protocol. As I predicted, opportunities have
arisen for further data collection and I have maintained ethical standards
in pursuing these. For example, I was invited to an early years conference
organised by the local authority and I collected artefacts for analysis and I
produced field notes of observations. Practical constraints have also
arisen. Due to the complexity of policy implementation, rather than one
children’s centre strategic partnership being in operation for me to
observe, I have observed a Community Partnership. I obtained informed
consent from the Partnership although it is loosely constituted and I
cannot be 100% certain that subsequent new members joining the group
are all fully informed about my research. However, I have taken every
opportunity to introduce myself, to provide my information sheet as well as
offering to withdraw on more than one occasion should the Partnership
wish me to. As I am maintaining strict confidentiality and anonymity I do
not believe that any harm could arise to these Partnership participants
4 Aug 09
23
whom I have observed. I have offered to feedback when I have interim
findings and am awaiting a response from the Partnership.
The letter … that approved my application explains that a substantial
amendment is that which is likely to affect:
(a) the safety or physical or mental integrity of the trial participants
(b) the scientific value of the trial
(c) the conduct of management of the trial.
Having reflected upon my research and discussed this in detail with
supervisors we do not believe that the inevitable minor changes to the
original protocol arising from fieldwork constitute a substantial
amendment.
I trust that this satisfies your monitoring requirements and I hope to
provide you with a final report in due course. “
I was informed in writing that my response was satisfactory so with relief I
proceeded and in subsequent annual reports reported no adverse incidents or
substantial changes. My final interaction with research governance has been to
send an “end of study” report and an executive summary of my thesis that I wrote
in a non-academic format.
7. Writing Up ethically
Ethnographers face ethical decisions over how to represent including how to
select data from the voluminous mound that qualitative research often generates.
(Mason, 2004). The ethical issue of bias entails complex philosophical questions
that are not addressed by the LREC questions about conflicts of interest. Here I
want to discuss the ethics of analysis, bias and the perils of overclaiming. There
are ethical dilemmas around how to interpret and reflect reality and how to derive
4 Aug 09
24
meaningful conclusions from what often feels like an overwhelming messy mass
of qualitative data. As my research design crystallized it became apparent that I
would be analyzing broad questions of partnership and network governance and
situating individuals as members of these institutional arrangements. Rather than
analyzing the minutiae of conversations, I focussed on discursive practices. 9 To
view research respondents, participants or subjects as bearers of discourse de-
centers subjectivity and this in itself might be viewed as unethical. Researchers
emphasising the agency of research respondents grant them the respect of
having a point of view that may be separate and independent of the researcher.
As I was seeking to explore power relations inherent in policy however, my
ethical stance or point of view meant that I analysed the dialectical processes of
structure and agency through which subjectivity was dialectically reproducing
policy. Following Katz (2002) I prioritised what I regarded as “luminous moments”
in my data. I was biased towards data that highlighted what I saw as policy
contradictions, such as the strategy of relying on the private childcare market to
solve the “wicked issue” of child poverty. I presented data on the expedient
practice of spending up to maximum budget by financial year end, irrespective of
whether such expenditure was directed towards satisfying childcare needs.
When I presented a work in progress paper based on this data, people in the
audience asked me directly whether I was claiming that public money was being
wasted. I faced the ethical dilemma of how to represent expedient allocation of
public funds (which I saw as being against the public interest, and certainly in
contradiction to the idea of carefully judged evidence based practice) without
claiming that the people I observed were acting unethically. To conclude that
public money was being wasted would be to directly attribute blame to the
agency of individuals. On the other hand, to negate their agency and interpret
waste as a fault of “the capitalist structure” or “the bureaucratic system” would
seem to be an unwarranted claim in terms of the data I had. Constructing an
argument for my thesis and theorizing helped me to gain a critical distance on the
data. Clarke J. (2004) pursues an analysis with doubts and hesitancies, noting
9 Except in Carter 2009 I do analyse the production of an inter-view
4 Aug 09
25
with frustration the tendency of some researchers to set out their analysis of how
power is all pervasive, relegating relative autonomy or resistance to an endnote.
Willis and Trondman argue that researchers should be willing to be surprised by
fieldwork and should remain open to what they term “a-ha” moments of
illumination. In their view, researchers should combine data and theory so that
“Engagement with the “real” world can bring surprise to theoretical formulations”.
(2000: 399). Mason (2004) urges caution with respect to claims of qualitative
research. She argues that, generally sociologists are only able to assert modest
claims as causality is rarely as clear cut in social science as it is in natural
sciences10.To return to my previous point about ethnography’s reliance on
rhetoric, I negotiated the ethical dilemma through argument and evocation in my
writing. Theoretically I drew on Lipsky’s work (1980) on street level bureaucrats
to locate the social actors structurally in relation to their limited, relative autonomy
vis-a-vis policy and I adopted Chenail’s advice to “let the data be the star”.
Chenail (1995:3)”. He argues that
“… the process of establishing the trustworthiness of any study comes
down to the quality of the relationship built between the researcher and
the community of readers and critics who examine the study. ” (p2)
By foregrounding data on expedient spending practices and juxtaposing verbatim
quotes about the forthcoming financial year end against statements about child
poverty, I was aiming for a representation of reality whereby I might evoke in the
reader an appreciation of the complexity of policy implementation but also allow
them to perceive a tragic sense of waste. This “thick description” is a key
strength of ethnography (Clifford and Marcus, 1986). Van Maanen (1995:201)
has discussed at length how ethnographic texts diverge from the “logico-scientific
code” of conventional social science writing. He notes how, in the “scenic
method”:
10 I am aware of debates about the uncertainty principle in physics but I am making a general point about probability and causality
4 Aug 09
26
“The writer shows rather than tells. Situations are recreated for the reader,
so that he or she can see and hear, smell and touch, listen to the
dialogue, feel the emotional tone.” (1995:118).
The RGF (DH, 2005) carries no recognition of the ethical dilemmas entailed in
writing up qualitative research nor of how emotions might be affected by reading
research accounts other than to acknowledge that reports may be written in
different formats for different audiences.
6. Conclusion
This paper has presented empirical data to show the extent to which my PhD
research was governed by the Local Research Ethics Committee. The committee
sanctioned my research and gave me ethical approval which I was used on
occasion in the field as a legitimation device in a perfunctory fashion. Although I
was required to comply with a two-stage informed consent process for interviews,
this did not guarantee ethical practice as my confession and discussion of the
ethics of transcription shows. Despite concerns about competing epistemological
paradigms (Hoeyer, 2005) I found in practice that I did have some opportunity to
justify my critical theoretical approach and iterative research design. I took some
risks in pursuing covert research and was honest about how I considered the
potential ethical harms of this but weighted them against research benefits. To
some extent my ethical behaviour was monitored by the committee querying a
vague statement in my annual report. On the other hand, what I regarded as a
serious adverse event (causing pain and discomfort to two interviewees)
occurred beneath the radar of the committee’s oversight function. The committee
cannot possibly know whether I have maintained confidentiality. Aside from my
confession, the ethical negotiation of trust between myself and those I observed
and / or interviewed happened despite, rather than because of research
governance. Had I complied strictly with the recommendation and disrupted
4 Aug 09
27
policy implementation business by withdrawing from meetings to seek informed
consent, I doubt I would have maintained credibility or rapport and would not
have been able to gather rich, naturalistic data.
I argue that the codified ethics of the Research Governance Framework are not
the same as personal values and moral practice. Whilst I accept that these are
also socially constructed, my paper argues that an ethical sensibility is beyond
codification, located in the infinite contingency of principled praxis and situational
ethics (Mauthner et al , 2002). Ethnography does not deserve to be immune from
ethical considerations and I do not call here for “… denying the oxygen of
legitimacy to the self-appointed sanitary inspectors of HSS” [humanities and
social sciences] (Dingwall, 2008:11). In the short term there must be further
dialogue so that ethics committee decision makers develop a greater awareness
of how their procedures impact (or don’t impact) on qualitative social science and
so they recognise and value the potential of ethnography to get a close up
perspective on ethical dilemmas as they occur in policy, in research and in
practice.
bibliog. 943 word count 4/8 8456
References Aldred, R. (2008) “Ethical and Political Issues in Contemporary Research
Relationships” Sociology 2008; 42; 887-903
Alldred P and Gillies V Eliciting Research Accounts: Re/Producing Modern
Subjects? In Mauthner et al (2002) Ethics in Qualitative Research London : Sage
Publications Ltd.
4 Aug 09
28
Les Back (2002) 'Dancing and Wrestling with Scholarship: Things to do and
things to avoid in a PhD Career' Sociological Research Online, vol. 7, no. 4,
<http://www.socresonline.org.uk/7/4/back.html
Bailey, R. (2001) "Overcoming Veriphobia- Learning to Love Truth Again", British
Journal of Educational Studies, 49 (2) pp. 159-172.
Barrett S (2004) “Implementation Studies Time for a Revival? Personal
Reflections on 20 Years of Implementation Studies” Public Administration 82 (2)
Bently, C. and P. Enderby (2005) ‘Academic Medicine: Who is it For?’ British Medical Journal 330: 61 British Sociological Association (2004) Statement of Ethical Practice
Bryman A (1988) Doing Research in Organizations London: Routledge
Duke K (2002) “Getting Beyond the ‘Official Line’: Reflections on Dilemmas of
Access, Knowledge and Power in Researching Policy Networks Journal of Social
Policy 31 (1) 39-59 Cambridge University Press
Buchanan, D., Boddy, D. and McCalman, J. (1988) “Getting In, Getting On,
Getting Out and Getting Back” in Bryman A. (1988) Doing Research in
Organizations London: Routledge
Calvey, D. (2008) “The Art and Politics of Covert Research: Doing `Situated Ethics' in the Field” Sociology 42 (5) Clarke, J. and Hall, E. (2008) “Will the lessons be learned? Reflections on local
authority evaluations and the use of research evidence” Evidence & Policy 4 (2)
255-68.
Clifford J. and Marcus G.F. (1986) Writing Culture The Poetics and Politics of
Ethnography University of California Press Berkeley
4 Aug 09
29
Corden, A. and Sainsbury, R. (2005) “Research participants’ views on use of
verbatim quotations” ESRC Research Methods Programme Working Paper No
23 York: Social Policy Research Unit University of York
DH (2005) Research Governance Framework for Health and Social Care
Dingwall, R. (2006) Confronting the Anti-Democrats: The Unethical Nature of Ethical Regulation Social Science Medical Sociology Online (1) Dingwall, R. (2008) “The ethical case against ethical regulation in humanities and social science research” 21st Century Society 3(1) Duncombe, J. and Jessop, J. “’Doing rapport’ and the ethics of ‘faking friendship’”
in Mauthner et al eds. (2002) Ethics in Qualitative Research London: Sage
Publications Ltd.
Edwards, R. ed. (2004) Social Capital in the Field : Researchers’ Tales London
South Bank University Families and Social Capital ESRC Research Group
Edwards R. and Mauthner, M. (2002) “Ethics and Feminist Research:Theory and
Practice” in Mauthner et al (eds.) book Ethics in Qualitative Research London:
Sage Publications Ltd.
ESRC (2005) Research Ethics Framework Fairclough, N. (1992) Discourse and Social Change Cambridge: Polity Press.
Fischer, F. 2003, Reframing Public Policy Oxford: Oxford University Press,
4 Aug 09
30
Foucault, M. (1980) Power/Knowledge Selected interviews and other writings
1972-1977 by Michel Foucault edited by Colin Gordon Harlow The Harvester
Press Ltd.
Gillies, V. and Alldred, P. (2002) “The Ethics of Intention: Research as a Political
Tool” in Mauthner et al eds. (2002) Ethics in Qualitative Research London: Sage
Publications Ltd.
Golden-Biddle, K. & Locke, K. (1993) “Appealing Work An Investigation of How
Ethnographic Texts Convince” Organization Science 4(4) 595-616.
Haas P. (2004) “When does power listen to truth? A constructivist
approach to the policy process” Journal of European Public Policy,
11(4)
Hannigan, B. D. Allen D. “A tale of two studies: research governance issues
arising from two ethnographic investigations into the organisation of health and
social care” International Journal of Nursing Studies 40 (2003)
Hedgecoe, A. (2008) “Research Ethics Review and the Sociological
Research Relationship” Sociology 42(5)
Hoeyer K., Dahlager L.& Niels L. (2005) “Conflicting notions of research ethics
The mutually challenging traditions of social scientists and medical researchers”
Social Science & Medicine 61 1741–1749
Katz,J (2002) From how to why On luminous description and causal inference in
ethnography (Part 2) Ethnography 3(1): 63–90
Levitas, R. (1998) The Inclusive Society? Social Exclusion and New Labour
Houndmills: Palgrave.
4 Aug 09
31
Lipsky, M. (1980) Street-Level Bureaucracy dilemmas of the individual in public
services New York: The Russel Sage Foundation.
Mason, J. (2004) Qualitative Researching London: Sage Publications Ltd.
Mauthner M., Birch M., Jessop J. and Miller T. (2002) Ethics in Qualitative
Research London: Sage Publications Ltd.
Miller,T. and Bell, L. “Consenting to what? issues of access, gate-keeping and
‘informed’ consent” in Mauthner et al (2002) Ethics in Qualitative Research
London: Sage Publications Ltd.
Murphy, E. and Dingwall, R..(2007) ” Informed Consent, Anticipatory Regulation
and Ethnographic Practice” Social Science & Medicine 65 2223–2234
Newman, J. (2005) Remaking Governance, Peoples, Politics and the Public
Sphere Bristol.: Polity Press
Newman, J. and Clarke, J. (2009) Publics, Politics and Power Remaking the
Public in Public Services London: Sage Publications Ltd.
Richardson,S. and McMullan, M. (2007) Research Ethics in the UK: What Can
Sociology Learn from Health? Sociology 41(6): 1115–1132
Sanger, J. (1996) The Compleat Observer? A field guide to observation London:
The Falmer Press
Schostak, J. (2006) Interviewing and Representation in Qualitative Research
Maidenhead : Open University Press.
Shaw S., Petchey, R. P.and Chapman, J., Abbott, S. (2009)double-edged
sword? Health research and research governance in UK primary care Social
Science & Medicine 68
4 Aug 09
32
Shaw, S. and Greenhalgh, T. (2008) Best research For what? Best health For
whom? A critical exploration of primary care research using discourse analysis
Social Science & Medicine 66 2506=2519
Social Policy Association Guidelines on Research Ethics
Spicker, P. (2007) “The Ethics of Policy Research” Evidence and Policy 3(1)
Stake R (2000) “The Case Study Method in Social Inquiry” in Gomm (2000)
Van Maanen, J. (1995) Representation in Ethnography London : Sage
Publications Ltd.
Willis, P. and Trondman, M. (2000) “Manifesto for Ethnography” Ethnography